Talk:Paper - Embryonic and postnatal development of the adrenal cortex, particularly the zona glomerulosa and accessory nodules
EMBRYONIC AND POSTNATAL DEVELOPMENT OF THE ADRENAL CORTEX, PARTICULARLY THE ZONA GLOMERULOSA AND ACCESSORY NODULES
Peter Gruenwald Department of Pathology, Children’s Hospital and Infants’ Hospital, Boston, Massachusetts
Three Plates (Eighteen Figures)
Recent investigations of the adrenal have suggested that new cortical tissue differentiates in the capsule of the mature gland, and subsequently becomes part of the zona glomerulosa (Zwemer, Wotton and Norkus, ’38; Bachmann, ’39; Gruenwald, 42a; Wotton and Zwemer, ’43; Gruenwald and Konikov, 44). A close relationship of the cells of the capsule to those of the cortex is further suggested by the histological demonstration of ascorbic acid (Heckel, ’42) and of vitamin A (by fluorescence microscopy, Popper, ’41) in the capsule. The mechanism of incorporation of newly formed cortical tissue into the cortex proper depends on its amount, and on the structure of the zona glomerulosa (Gruenwald and Konikov). It has also been claimed, that in this region, as well as at the inner border of the cortex, not only differentiation but also degeneration of cortical tissue may occur under the control of the corticotropic hormone of the hypophysis (Tonutti, ’41, ’42b). Thus, the examination of the capsule and zona glomerulosa may, after elucidation of the normal phases of their changes, become an important part of the study of the adrenal cortex under experimental and pathological conditions,
+ Author’s present address: Department of Pathology, Long Island College of Medicine, Brooklyn, N. Y.
In the course of an examination of large numbers of human adrenals it became apparent that the structure of the zona glomerulosa, and the form and amount of apparently newly formed cortical tissue, show considerable changes during infaney and childhood. The present work is concerned mainly with this aspect of cortical development. The structure of the outermost layers of the human adrenal and its changes from embryonic life to maturity will be reported, including findings in accessory cortical nodules. This work is limited to observations made with the usual histological methods applied to paraffin sections. Only in a few instances was it supplemented by the examination of sudanophil lipoid in‘frozen sections. No studies were made of evidence of endocrine and other specific metabolic activity of the cells by physical or histochemical methods, such as the demonstration of steroids, plasmal, or vitamins A or C. The present results are intended to provide a basis for further histochemical, experimental and pathological work, and are not expected to be final until confirmed bv these methods.
The following excerpts from the literature on adrenal development will serve as a starting point for the discussion of the observations. Many authors have examined the early development of the organ in man and mammals from the coelomic wall, and most of them have come to the conclusion that epithelial cords of the cortex primordium arise as outgrowth of the coelomic epithelium. A recent representative of this opinion, reporting on human embrvos, is Politzer (786). Goormaghtigh (’21) and Waring (’35) failed to find distinct epithelial buds, and more recently Uotila (’40) said of the cortex primordia in human embryos that ‘‘the formation of distinct cell buds has not been observed.’’ In a previous investigation, the present author (’42a) also failed to find epithelial cords forming the early cortex primordium. The cell mass from which the cortex arises is purely mesenchymal in its structure for a considerable time after its formation, as can easily be demonstrated by impregnation of lattice fibers surrounding each cell. The cells are part of the mesenchyme of the coelomic wall, and many of them derive from dividing cells of the coelomic lining in much the same way as do many cells of the nearby connective tissue. The lining of the coelom is not epithelial itself at this time and place, and when it gives rise to cells of the cortex anlage, it does so without the formation of epithelial cords. It was also pointed out that, judging from the presence of lattice fibers enveloping single cells, the greater part of the adrenal cortex permanently remains non-epithelial. Only a peripheral layer, roughly corresponding with the zona glomerulosa, gradually loses the argyrophil fibers within its cell cords (Gruenwald and Konikow), and is then truly epithelial. Most adrenals show fluent transitions between epithelial and non-epithelial structure at the border of the two portions of the cortex.
Much has been written about that part of the adrenal cortex of the human fetus which degenerates shortly after birth, and is usually referred to as fetal cortex or x-zone. The former name is not entirely correct because the adrenal cortex of the fetus contains the permanent as well as the degenerating portion. In view of our meomplete knowledge of the nature of the latter portion, the term x-zone appears most appropriate, Time and manner of its degeneration have been thoroughly studied by Lewis and Pappenheimer (716) and Benner (’40). Only a few mammals have an x-zone, among them the mouse, and in this species differences were found between various strains. The x-zone of the mouse has been subjected to experimentation, and from its reaction to hormonal influences Gersh and Grollman (’39) concluded that it is not essentially different from the rest of the cortex.
Age differences in the postnatal human zona glomerulosa have, to the best of the author’s knowledge, not been investigated. Gruenwald and Konikov briefly described such differences in eats.
Many authors assume that there is a cell movement in the mature adrenal cortex from the periphery toward the zona reticularis. The literature on this subject may be found in the articles of Blumenthal (’40) on mitotie aetivity in the outer part of the zona fasciculata, and of Hoerr (’31) and Bennett (’40) dealing with the changing properties of cortical cells in various layers. Cell movement from the site of mitotic activity in the outer zona fasciculata toward the zona reticularis obviously does not concern the zona glomerulosa, and many authors have failed to consider the role of the latter zone in the life and function of the adrenal cortex. Mulon (’03) regarded the zona glomerulosa as the germinal layer, but his findings of frequent mitosis and amitosis in that zone have not been confirmed. Others ascribe to the zona glomerulosa the role of a cell reservoir. This has been discussed and older literature quoted by da Costa (’22, 383). A tentative concept, including this phase as well as the more recently discovered new formation of cortical tissue in the capsule, has been proposed by Gruenwald and Konikov (’44). According to this hypothesis, multiplication of cortical cells by mitosis in the outer fasciculata may be supplemented or replaced by the release of cells from the reserve store in the zona glomerulosa. Depleted stores may in turn be replenished either by mitosis in the outer fasciculata or by apposition of newly differentiated cortical cells from the capsule. In support of the reservoir function of the zona glomerulosa, its transformation into part of the zona fasciculata has been described (Goormaghtigh, 718; da Costa, ’83; Gruenwald and Konikov, ’44), apparently under the influence of a demand for functional cells.
Several authors do not agree with this concept of inward movement of cells in the mature adrenal cortex. Whitehead (743) claims that mitosis occurs in the cortex solely during the growth period and not in adult animals, and that only growth of the cortex, but not continuous cell replacement, occurs by cell division. Blumenthal (’45) recently found a decrease, but no complete disappearance of mitoses even in old guinea pigs. Tonutti (’41, ’42) holds that growth as well as degeneration occurs simultaneously at the outer and inner border of the cortex, and that there is no cell movement toward the zona reticularis.
MATERIAL AND METHODS
The early development of the adrenal cortex in embryos of the second month has been described in a previous report (°42a). The same material, consisting of thirty-seven serially sectioned human embryos of 7.5 to 32mm, was again reviewed, mainly in order to examine the alleged separate origin of the permanent cortex by a second proliferation from the coelomic lining (Keene and Hewer, ’27; Uotila, ’40). The sections of these embryos are stained mostly by the azan method; in a smaller number of specimens impregnation of lattice fibers (method of Gomori) was used. Adrenals of ten human embryos of 6.7 to 22 cm crown-heel length were also available. Jn each case at least part of one organ was sectioned serially and various stains were employed, among them in all cases, the azan method and silver impregnation. _
Postnatal material was selected from a large number of routine sections of organs fixed in Zenker’s fluid, and new sections of suitable specimens were cut and stained by various methods. The age distribution of the postnatal material is as follows:
Newborn to 1 day ............ 8 ldaytolmonth .............. 12 lto6months ...............4. 19 6to12 months ................ 8 lto2 years ......... 00. eee 10 2to3 years ..........4.e eee 8 3tolO years ............0006. 15 10 to 20 years 2.12... 2. eee 9 20 to 78 years ..... ce ee ee ee 24
Organs of several premature infants are included in this material.
Selection of proper staining methods proved to be of great importance in the present work. The usual hematoxylin and eosin stain does not show adequately the form and arrangement of the cortical cell groups. It is necessary to stain the connective tissue fibers which bind these cell groups. Stains for collagenous fibers are inadequate, particularly in organs of embryos and infants which are devoid of collagenous fibers, except in the densest portion of the capsule and the walls of larger blood vessels. Impregnation of lattice fibers gives an excellent view of the architecture of the adrenal cortex, even in those cases in which many other stains show but a uniform mass of cells, such as in the outer portion of the fetal adrenal. In addition, other stains were used to show the cells themselves. A combination of any of these stains with silver impregnation to the extent of showing much cellular detail has not been used because fine lattice fibers are then obscured.
Embryos of the second month, Previous work on the origin of the adrenal cortex primordium was reviewed above, ineluding the author’s own findings favoring the mesenchymal origin of the anlage (’42a). According to Keene and Hewer (°27) and Uotila (’40), a second proliferation of the coelomic lining takes place in embryos of about 12mm, and the cells thus formed (the epithelial cap of Keene and Hewer) surround the first anlage and form the permanent cortex. Hxamination of embryos at this age shows, particularly with the use of silver impregnation, that the adrenal primordium contains no epithelial cell groups. The cortical cells are swollen cells of the mesenchyme, and it is difficult if not impossible to distinguish the superficial cortical cells from the adjacent mesenchyme at all points. Proliferation of the coclomie lining still oceurs at the stage designated by the above mentioned authors but, just as in the earlier phase, not in the form of epithelial sprouts. Instead, more mesenchyme is formed in the adrenal region, and it is probable that some of the cells partially surround the original anlage and then differentiate into cortical cells. It is difficult to obtain reliable information on the amount of cortical tissue thus formed, and the impression just described was obtained by judging the distance of the cortex primordium from nearby structures and thus the growth of the organ at the expense of the mesenchyme. Nothing was seen to confirm the alleged migration of cells derived from the coelomic lining around the entire cortex primordium.
These observations lead to the conclusion that the adrenal cortex develops by rapid differentiation of mesenchyme of the coelomic wall during the early phase and a slow continuation during a subsequent phase, rather than a succession of two separate proliferations. No epithelial cell groups are present in the primordia during this and a long successive period (fig. 1).
When proliferation of mesenchyme from the eoelomic lining and differentiation of cortical cells from the mesenchyme cease, the adrenal is composed of swollen cells which are large and have a more acidophilic cytoplasm in the central portions than in the periphery. Each cell is surrounded by a network of lattice fibers. Wide blood vessels make their appearance in the central part of the organ and give it a spongy structure while the distal zone remains solid (fig. 1). Gradual transitions lead from large to small cells, and from spongy to solid arrangement. It is impossible at any stage of intrauterine life to draw a sharp line between cells which differentiated at different times in the early embryo, or between x-zone and permanent cortex, or any other lavers in the cortex. Just as in the mature organ, gradual transitions are seen everywhere, due perhaps to gradual changes in the differentiation of cells traversing the zones. It is, therefore, unjustified to speak of separate anlagen of the x-zone and the permanent cortex.
Later embryonic stages. Tn the course of several months of embryonic life, two zones joined by an area of transition develop in the adrenal cortex. (The temporary presence in the cortex of groups of migrating cells of the future medulla will not be described.) Lattice fibers separating each cell from its neighbor disappear in the outer zone of small cells (fig. 2; for an early stage of this change in the dog embryo see Gruenwald and Konikov, fig. 23). The difference in cell size between the inner and outer zone becomes more marked (fig. 3). As a result, the outer zone consists of large, solid, epithelial cell masses, subdivided by blood vessels and lattice fiber trabeculae at a great distance from one another. On section, the width of these cell masses from one connective tissue partition to the next varies considerably. In some areas no such partitions are present throughout an entire lowpower field or a large part of one, while in other regions many more subdivisions may be found. There seems to be a trend toward division of larger cell complexes in the outer zone into smaller ones during late embryonic life.
This outer layer of the embryonic cortex has been considered either as the zona glomerulosa or as the permanent cortex, in which case it would correspond to both the zona glomerulosa and the permanent zona fasciculata of the postnatal gland. It was pointed out above that the existence of fluent transitions between zones makes any comparison of layers at different stages very unreliable. When the x-zone degenerates after birth some of the large cells always remain intact and eventually become part of the permanent zona fasciculata. Comparative considerations are also in favor of the origin of the permanent zona fasciculata from the large cells of the embryonic gland because these cells are also present in the embryos of those species which have no x-zone that degenerates after birth.
The well known cavities appear in the outer zone of the embryonic adrenal shortly after it has acquired its epithelial structure. Their size and number vary greatly. Figure 3 shows an example of an organ with numerous cavities. Their morphogenetic importance is not understood. We do not know whether their appearance is an essential part of cortex development, or an unimportant event. Hett (’25) favors the first mentioned view, and believes that the formation of the zona glomerulosa with its arches and cell balls depends on these cavities. Many other workers have described them without attributing to them much importance for the future development of the gland. Da Costa (’28) sees in the cavities an expression of a fundamental glandular structure of the cortical cell cords. In human embryos the cavities show considerable variation in number and are present for a varving length of time. In some cases they persist after birth for many months (fig. 8), and similar cavities have been found in adult mammals (Kolner, 718; Kohno, ’25; Da Costa, ’28). Formation of the zona glomerulosa, on the other hand, proceeds in human embryos simultaneously all over the surface of the organ. This, as well as the presence of cavities long after differentiation of the zona glomerulosa, speaks against Hett’s assumption of an important role of these spaces in the development of the zona. As has just been described, the zona glomerulosa develops by a slow process of differentiation characterized by the disappearance of lattice fibers surrounding: single cells, the appearance of a characteristic difference of cell size and staining properties between these cells and those of the deeper zona, and the formation of large cell masses connecting with many of the slender cell cords of the deep layer.
Postnatal. development. It is difficult to describe definite stages in the postnatal development of the adrenal cortex because of a wide variation not only in the time at which certain changes occur, but also in the stages of development found in adjacent regions of the same gland. The only change in the infant’s adrenal which shows a fairly close correlation with age is the degeneration of the x-zone, and this will not be deseribed here. In presenting descriptions and photographs of parts of adrenals, it will always be understood (unless otherwise indicated) that the condition is typical of the given stage but not the only one to be seen in that organ, and that it may occur occasionally in younger glands and frequently in older ones.
Newborn to 6 months. The adrenal cortex of the newborn consists of the two lavers which were described above. The greater part of the inner layer of large cells forms the x-zone and degenerates during the first few months of extrauterine life. Its demarcation from the permanent cortex is not a sharp one until the process is far advanced; the difficulties in establishing the identity of the x-zone with any part of the fetal organ have been referred to above.
At birth the zona glomerulosa consists of the wide arches found in the fetus. It is difficult to visualize the shape of these units even with the aid of serial sections. They are probably irregular plates extending parallel to the capsule and connecting on their inner side with the ends of large numbers of fasciculata cords (fig. 4). Persistence of wide undivided glomerulosa units resembling those of the fetus was seen in a 2} months old infant in an otherwise normal adrenal (fig. 6). The mechanism of subdivision of large units into smaller ones is not apparent from the study of the present material. In one instance the appearance of a lattice fiber framework within large glomerulosa units suggests a change of differentiation and arrangement (fig. 5), producing narrow glomerulosa cords, while heavier fibers still indicate the boundaries of the old units.
There is also great variability in the height of the glomerulosa units from the capsule to the level at which lattice fibers appear between the cells (compare figs. 4 and 6). Tn voung infants the zona glomerulosa consists of small cells with dark staining nuclei as described above in fetal glands. There is a gradual] transition to larger cells with a very pale staining cytoplasm and somewhat larger nuclei, giving the entire Javer a pale appearanee when stained with the usual methods. This change may occur at any time between the ages of 2 weeks and 6 months. It is preceded by a gradual decrease in width of the zone of small cells, and does not affect all cell groups simultaneously so that dark and pale staining glomerulosa units may be present side by side (fig. 7). Sudan-stained frozen sections show numerous small lipoid droplets in the pale staining glomerulosa units. The change just described is preceded by the disappearance of the cavities which are present in varying numbers of glomerulosa units of vounger organs. The oldest gland in which numerous cavities were seen, at the age of 6 months, is shown in figure 8. In this case, however, debris is present in many of the cavities, suggesting that they may not he strietly comparable with fetal cavities.
Accessory cortical nodules in the vicinity of the main gland are found very frequently in infants. According to the review of Dietrich and Siegmund (’26) they are present in almost all individuals during the first year and decrease in number during childhood. In the present material, adrenals of thirtyfour infants up to 3 months of age were examined in single routine sections, and twenty-four of these sections showed accessory nodules. In most cases multiple cell groups were found in or near the capsule. It is probable that most of the remaining ten cases also had accessory nodules which were not cut in these single sections. Figure 10 shows the case in which the largest number of accessory cortical cell groups was present,
Accessory nodules appear in various forms and varving relations to the main gland. Often they are spherical and, if large enough, show the same zones as the main gland in corresponding stages of their development. This includes degeneration of the central portions of the nodules at the time when the x-zone degenerates in the adrenal proper. Many of these nodules contain in their own capsule cords or sheets of cortical cells (fig. 12) which closely resemble those present in the eapsule of the main glands, only in later stages. That suggests that accessory nodules may grow by apposition of newly differentiated cords at a time when this is uncommon in the main gland. Thomas (’11) described this long before the discovery of differentiation of cortical cords in the capsule of mature adrenals. He assumed that cortical cells had split off from the primordium of the adrenal in the early embryo, and remained inactive in the capsule until some unknown stimulus eaused their differentiation. Since we assume that cells of the capsule and nearby connective tissue have the potency to form cortical tissue, the assumption of cells splitting off from the embryonic adrenal has become superfluous. Accessory nodules far from the main gland develop like the ones in the capsule. Figure 18 shows a nodule in the epididymis of a 2 months old infant with cortical cords in the surrounding connective tissue and far advanced degencration of the x-zone in the center. 402 PETER GRUENWALD
Occasionally the capsule of the adrenal proper contains groups of cortical cell sheets resembling those in older children and adults, which are believed to replenish the cortex by apposition (Zwemer, Wotton and Norkus, 738; Wotton and Zwemer, ’43; Gruenwald and Konikov, 744). However, these formations in young infants differ from those of older individuals by forming compact groups which reach a considerable size and differentiation before a narrow connecting bridge to the main gland is established (fig. 11). In this respect they resemble accessory nodules rather than stages of apposition. Single cords joining the cortex, as in accessory nodes or in the main glands of older persons, do not occur in the adrenals proper of young infants.
Accessory cortical cell masses in the capsule may be seen in all stages of fusion with the main gland. This may be interpreted as indicating either separation from the main gland, or independent origin and secondary fusion with it. Stecksén (’02) favored the former alternative. Against this is the decrease in the number of accessory nodules with advancing age (see above), and the fact that growth of accessory nodules (and in later stages also the main gland) by apposition can easily effect a fusion if it fills the space between the nodule and the adrenal proper. In those cases in which the accessory tissue is present in the form of cords and sheets ot cortical cells spread between the connective tissue layers of the capsule, outgrowth from the main gland cannot very well be assumed; it is very doubtful that normal cortical tissue should be capable of this extent of infiltrative growth. Degeneration of cortical nodules by extreme vacuolation and fatty changes of their cells (Stecksén) has not been found in the present material. Perhaps lobules of immature fat tissue with foamy cells, which are often seen near the capsule of the adrenal in infants, were mistaken in the past for degenerating cortical tissue.
The old idea of the origin of accessory cortical nodules by deviation of epithelial sprouts in the early embryo was based on the assumption that the normal cortex itself is derived from epithelial cords growing in from the peritoneal epithelium. It was mentioned above that this is not the case. There is reason to believe that the potency to form cortical tissue is not limited to the cells which do so in the early embryo. Evidence of this is the differentiation of cortical cell cords in the capsule of the mature organ, as well as the recently discovered development of functionally adequate cortical tissue in the ovaries of ground squirrels (Groat, 43, ’44) and mice (Hill, 46) after adrenalectomy. In view of these facts it appears most probable that accessory cortical tissue forms by differentiation in loco of cells of the connective tissue which are normally endowed with the necessary developmental potency (Gruenwald, ’42b).
The accessory nodule undergoes a characteristic change in the course of what is believed to be its junction with the adrenal proper. It is incorporated into the structural pattern of the main gland in the sense that its cell cords show transitions from glomerulosa to fasciculata structure no longer in the direction of the center of the nodule, but toward the center of the main gland. Even if there is only a narrow connecting bridge, fasciculata cords traverse it and continue from the nodule through a breach in the zona glomerulosa directly into the fasciculata of the main gland.
One-half to 3 years. During this period the relative uniformity of the peripheral portion of the adrenal disappears. Areas of widely different structure may often be seen in one section. The transformation of the zona glomerulosa into a layer of pale staining cells with a foamy cytoplasm continues, and in many areas silver impregnation reveals a peculiar separation of this zone from the underlying parts by a continuous layer of lattice fibers (fig. 9). These fibers are obviously not derived from the capsule, and their presence under a superficial zone of cortical tissue cannot, as in the adult, be regarded as an indication that the outer zona has recently differentiated in the capsule. A similar separation of the zona glomerulosa from the fasciculata by lattice fibers has been found in adult cats, and in a striking form in cattle embryos (Gruenwald and Konikov, ’44). Its significance is unknown. Frozen sections of the adrenal shown in figure 9 were stained with Scharlach R. The sharply delimited zona glomerulosa is very rich in lipoid, more so than the fasciculata.
Tn other parts of adrenals of this age group, apposition of newly formed cords has apparently begun or may begin at any time. These areas are found most commonly near the edge of the leaflets of the gland, and show thin cell cords in the zona glomerulosa which are not thicker than those of the zona fasciculata into which they coutinue proximally. Their distal ends are often thinned and curved in such a manner that they are almost or entirely parallel to the capsule. The ends of these cords can often be found between the fiber layers of the capsule (fig. 14). This condition is identical with that described as suggesting apposition of newly formed cortical cords (Zwemer, Wotton and Norkus, ’38; Gruenwald and Konikov, ’44).
In addition to the large and often proximally delimited glomerulosa units and the thin, oblique cords which have so far been described, the adrenals contain during the period under consideration many cell cords in their zona glomeruloss which are less conspicuous. These are mostly somewhat wider than fasciculata cords, and arranged in radial direction. As development goes on, these cords increase considerably in number, until they dominate the field in the adult.
The accessory nodules do not differ essentially from those described in the preceding age group. They correspond in their structure with the main gland if their size permits the differentiation of zones. They are often surrounded by large numbers of apparently newly formed cortical cords. One difference is often apparent between the adrenal proper and its aecessory cell groups. The extensive accumulation of lipoid droplets in the cells of the zona glomerulosa is not duplicated in the accessory nodules. The glomerulosa cells of these nodules usually contain less lipoid than does their zona fasciculata.
Three to 10 years. All forms of glomerulosa units of the preceding age group may be seen in adrenals of children up to 10 years and older. In addition to these, a new form is comparatively frequent and often appears in large areas of the zona glomerulosa. Figure 15 shows a large portion of a leaflet from an adrenal of a 6-year-old child. Arches of cortical tissue resembling the zona glomerulosa of younger children are seen here not on the surface of the gland, but separated from the capsule by a conspicuous layer of thin cortical cords in an irregular arrangement. In some portions the superficial layer is thin, and the cords in it are arranged roughly parallel to the capsule (fig. 16). This suggests that the entire peripheral layer consists of newly differentiated cords laid down over the old zona glomerulosa. The other alternative, namely, outgrowth of these cords from the zona glomerulosa or development by transformation of the latter, does not fully explain the existence of wide and apparently undisturbed units of the zona glomerulosa in the depth. The relative frequency with which this condition is found indicates that it either persists with little change for a long period of time or occurs repeatedly.
In addition to accessory nodules similar to those previously described, a new form of cortical cell masses may now be seen (fig. 17). These are areas within the cortex of the main gland which are limited against the surrounding tissue either completely or only distally and on their sides. The peripheral cell groups in these regions have the character of zona eglomerulosa even though they may be located at the level of the zona fasciculata. In their arrangement they appear to be part of a separate nodule. An explanation of the origin of these nodules is suggested by the fact that they were not seen in infants, and by the appearance of some of them in serial sections. They may then be seen to emerge on the surface of the organ and have the typical appearance of aecessory nodules fused with the main gland. All this suggests that they are accessory nodules which have been partially taken up into the main gland. However, only a small proportion of the accessory nodules retain this individuality after their fusion with the adrenal proper.
Ten years through maturity. Histological evidence of apposition of newly differentiated cortical tissue from the capsule of the mature adrenal has been quoted above. The present description is intended merely to establish a link between the preceding sections and the better known condition in the adult. Whatever changes occur in the outer zone of the adrenal after the tenth year can no longer be broken down into stages associated with limited age periods because the great variability makes such classification impossible. The changes described in the preceding sections, such as division of large glomerulosa units into smaller ones, the appearance of cords radiating obliquely into the capsule, and the formation of new layers of cortical cords over the old zona elomerulosa, all tend to produce a glomerulosa consisting of thin cell cords. These form direct continuations of the fasciculata cords without a conspicuous change in caliber. Wide cell columns in the zona glomerulosa are relatively uncommon in mature adrenals. In some areas many thin cords radiate toward the capsule at various angles, joining apparently newly formed cords or offering opportunities for such junctions. In some instances these groups of cords have the shape of a wedge with the base at the capsule (fig. 18, left half and extreme right), and other groups of cords with no apparent relation to the process of apposition may be seen between these wedges. The condition produced by large scale apposition of cortical tissue differentiating in the capsule has been described elsewhere (Gruenwald and Konikov, ’44).
The present material is not sufficient to study changes which may appear in the periphery of the adrenal in old age. This subject will therefore not be discussed.
For a proper understanding of the morphology of the adrenal cortex and its changes throughout life, it is important to remember that the organ develops from a mesenchymal DEVELOPMENT OF THE ADRENAL CORTEX 407
primordium, the cells of which are derived from the wall of the coelom including its mesothelium as well as the mesenchyme. For a long period of time the embryonic adrenal cortex remains entirely non-epithelial. Only during the fourth month of intrauterine life the cells of the peripheral zone form epithelial aggregates, as may be demonstrated by the lack of lattice fibers surrounding single cells. The transition between the epithelial and non-epithelial zones remains gradual throughout life with the exception of a temporary demarcation which may appear during childhood.
Collagenous connective tissue fibers are absent from the adrenals of infants except in the eapsule and the walls of larger blood vessels. It is, therefore, imperative to impregnate the lattice fibers in order to study the architecture of the adrenal cortex. In the adult, collagenous fibers may reach from the capsule into the cortex for a short distanee between the glomerulosa units. Conspicuous bundles or layers of collagenous fibers, extending parallel] to the capsule within the cortex, have been interpreted as originating from the capsule and displaced inward by the growth of newly differentiated cortical cell masses.
Embryonic development as reviewed and partly re-examined above produces a condition which is characterized by relatively large glomerulosa units (as compared with the adult) and the absence of forms suggesting apposition of cortical cords newly differentiating in the capsule. During childhood the zona glomerulosa gradually acquires small units which are in most instances direct continuations of single fasciculata cords, At the same time, indications of apposition of cortical tissue from the capsule begin to appear. <A relatively common condition in adrenals of older children is tentatively interpreted as a thick layer of newly differentiated cortex laid down over the old zona glomerulosa. Later on, as the appearance of the zona glomerulosa becomes less regular, signs of apposition appear in almost all glands in scattered small areas of the surface.
In a previous study (Gruenwald and Konikov, ’44) it was shown that those manmalian adrenals which have a well developed zona glomerulosa with large units usually show less apposition than others with an inconspicuous zona glomerulosa. On the assumption that the zone is a reserve store of cortical cells, this may be interpreted as indicating that organs with a large store need supplies of cells from the outside less frequently than others. The same relationship prevails in human adrenals of varying age; infants’ adrenals have a well developed zona glomerulosa and show little apposition of cortical tissue from the capsule. In the adult, on the other hand, the glomerulosa units are small and forms suggestive of apposition are common. The eat is the only species in which the adrenals of young and adult animals have been compared with regard to apposition of cortical tissue from the capsule and the conditions are the opposite of those in man. Young kittens show signs of abundant apposition and a poorly developed zona glomerulosa, whereas mature cats have a larger glomerulosa and, in the author’s material, never showed evidence of apposition (Gruenwald and Konikov). It is tempting to relate this difference to the abundance of cortical tissue in the human newborn. Perhaps the hormonal environment of the fetus produces a great excess of this tissue in man and not in the cat; large enough to leave a sufficient margin for the first period of postnatal life even after the degeneration of the x-zone.
An important part of the problem of apposition remains unsolved and that is the time factor. As long as we do not know the speed with which new cortical cells differentiate and become completely incorporated in the main organ, histological findings do not show the actual amount of apposition. This is of great interest, particularly in those adrenals of children which show what is here considered as evidence of very extensive addition of cortical tissue on top of the zona glomerulosa.
Whereas most authors who assume any cell movement in the adrenal cortex at all, conceive of it as proceeding toward the center of the organ, Tonutti has recently advanced a new concept. According to him, differentiation as well as regression of cortical tissue may occur at the boundary of cortex and capsule and simultaneously in the zona reticularis. It is true that the opposite processes of differentiation and degeneration may produce similar forms if one can only examine the condition prevailing at one moment. Yet, in the present findings, the assumption of apposition seems to be better justified, particularly where a zona glomerulosa is seen undisturbed under the layer in question.
Perhaps Tonutti’s new views will in the future help explain a condition in children which has no place in our present concept of adrenal architecture and growth. This is the separation of large areas of zona glomerulosa from the underlying cortex by a continuous layer of lattice fibers. It seems that the zona glomerulosa may separate itself temporarily from the fasciculata under conditions not yet understood. Similar observations have been made in cattle embryos and mature cats. Perhaps the zona glomerulosa can, under certain conditions, exist independently, as has been suggested on other grounds by Williams (’45).
It is generally assumed that the endocrine function of the adrenal cortex is related in some manner to the lipoid inclusions in its cells. What conclusions regarding the production or storage of the lipoid-soluble hormone can be drawn from simple fat stains, or various histochemical and physical methods, is an open question. Dosne and Dalton (’41) hold that accumulation of lipoid in cortical cells is in some instances an indication of storage rather than production of hormone, and that very active glands may be depleted of most of their lipoid. A small number of fat stains performed in the course of the present work have again shown the known fact that the zona glomerulosa of children is often very rich in lipoid — more so than the fasciculata. A similar distribution is oecasionally found in adults and in various mammals. If, as Sarason (48) claims, the ordinary sudan stain of frozen sections gives roughly the same results as various more complicated reactions which have been devised to demonstrate certain substances probably related to the specific function, one would have to assume considerable endocrine activity of the zona glomerulosa. This, however, is not in agreement with the concept that the outer portion of the zona fasciculata is the site of hormone production, and the glomerulosa a ‘‘presecretory’? zone (Bennett, ’40) or a reserve store of cortical cells. Perhaps this discrepancy is explained in part by the use of different species by various workers. Also, there is a definite possibility that not all lipoid inclusions in cortical cells have the same value in regard to endocrine function. This is suggested by a difference in the distribution of sudanophil lipoid, ascorbic acid, plasmal, or vitamin A (Fink, ’41; Kroezek, ’41; Popper, 741).
The pathological significance of cavities within cortical cell cords has recently been emphasized, and previous literature reviewed by Rich (’44). Other cavities which are normally present in the adrenals of human embryos and infants and also in certain adult mammals (da Costa, ’28) resemble these pathological cavities to some extent. However, pathological cavities often contain remnants of disintegrating cells which, according to Hett (’25), is never the case in normal embryos. The present specimens show, in accordance with Hett’s findings, no cells in embryonic eavities but a fine coagulum. There are, however, adrenals of infants in which it is difficult to decide whether cavities in the zona glomerulosa are of the normal or the pathological type. Tt has heen mentioned that such cavities may correspond to normal ones in their distribution and form but contain cells or debris (fig. 8). It might be of interest in the future to study the two kinds of eavities in their mutual relations.
Accessory cortical cell masses in the vicinity of the adrenal range from groups of cell cords interspersed between the fiber lavers of the capsule, to well organized and approwimately spherical nodules. The latter resemble the main gland in some respects and differ from it in others. Zones comparable to those of the main gland are present in nodules of sufficiently large size, and degeneration of the x-zone proceeds in the nodules at the same time as in the adrenal itself. Two differences were noted between accessory nodules and the main gland. One concerns the distribution of lipoid. The accessory nodules contain less lipoid, particularly in those cases in which the zona glomerulosa of the main gland is very rich in lipoid. The other difference is apparent only during the first year, and concerns the apposition of cortical cell cords which have newly differentiated in the capsule. This process apparently occurs frequently in the connective tissue surrounding aecessorv nodules, but not in direct relation to the main gland until later in childhood. It is not known whether the presence of numerous accessory nodules in an infant increases the ability to replace or inerease cortical tissue if needed. Even in the adult accessory nodules may behave differently from the main gland, particularly in pathological states. With regard to Addison’s disease, Kovaes (’28) and Guttman (’30) hold that cortical nodules may remain undamaged and hypertrophy to a limited extent and maintain life. Rogoff (’45) found in experiments with ligation of adrenal vessels that existing accessory nodules hypertrophy in some, but not in all cases, and supply part of the hormonal requirement.
There are indications that many accessory nodules in the eapsule join the main gland when the separating laver of connective tissue is pierced by differentiation of cortical tissue in it. In most instances the nodule loses its Independence as soon as it connects with the main gland, as is shown by the disappearance of the zona glomerulosa in the area of contact, and the continuation of the zona fasciculata of the one into that of the other. This is perhaps due to a diversion of the drainage of blood from the nodule toward the main gland as soon as the two come in contact, resulting in a rearrangement of capillaries and cortical cords. In a few eases, however, the nodule seems to retain some degree of individuality after joining the adrenal proper, and forms a distinct group of cords within the cortex. This is not believed to be the only 412 PETER GRUENWALD
manner in which distinct nodules within the main gland can arise.
The development of the adrenal cortex from a mesenchymal primordium that was derived from the coelomic lining, as previously described, has been confirmed, and some details have been added. The gland remains entirely non-epithelial until differentiation of the zona glomerulosa begins. This layer is the only one ever to acquire a truly epithelial structure,
At least a large part of the retroperitoneal mesenchyme appears to have the potency to form cortico-adrenal tissue. By differentiation in loco it gives rise not only to the cortex proper in the early embryo, but also to cortical cells in the adrenal capsule and to accessory nodules in later stages.
The zona glomerulosa of the older embryo and the infant consists of relatively wide and tall units in the form of plates or arches, each connecting with many fasciculata cords. These large units are gradually replaced by narrow cords, cither by subdivision of the existing units or by apposition of new fords over the old zona glomerulosa. Thus, the well known structure of the mature adrenal evolves; the cords of the zona eglomerulosa are mere distal continuations of fasciculata cords and seldom much wider than the latter.
Differentiation of cortical cell cords in the adrenal capsule, and their apposition to the gland itself, does not regularly occur in infants but is frequently seen in older children and adults. This is consistent with comparative results, in mammals, which have shown apposition less frequently in adrenals with a large, well differentiated zona glomerulosa than in those in which that layer is poorly represented.
Accessory cortical nodules in and near the adrenal capsule are a very common, if not a regular occurrence, in infants and voung children. They often show indications of active growth by apposition even before this process starts in the main gland. It is believed that many accessory nodules eventually fuse with the main gland.
The cavities which normally appear in the cell groups of the zona glomerulosa of embryos and infants are in some respects similar to pathological cavities in the zona glomerulosa and fasciculata of adrenals of all ages in the course of certain diseases.
The distribution of lipoid in the adrenal cortex is briefly discussed, and it is pointed out that the frequent occurrence of much lipoid in the zona fasciculata is not fully explained by our present concept of cortical morphology and function.
BacuMany, R. 1939 Zur Frage der Zona germinativa der Nebennierenrinde. Klin. Wochensehr., vol. 18, pp. 783-784.
BENNER, M,C. 1940 Studies on the involution of the fetal cortex of the adrenal glands. Am. J. Pathol., vol. 16, pp. 787-798.
Bennett, HE. 8. 1940 The life history and secretion of the adrenal cortex of the eat. Am. J. Anat., vol. 67, pp. 151-227,
BLUMENTHAL, H. T. 1940 The mitotic count in the adrenal cortex of normal guinea pigs. Endocrinology, vol. 27, pp. 477-480.
————— 1945 Aging process in the endocrine glands of the guinea pig. I. The influence of age, sex and pregnancy on the mitotic activity and the histologic structure of the thyroid, parathyroid and adrenal glands. Arch. Pathol., vol, 40, pp. 264-269,
Da Costa, A. C. 1922 Strueture ct histogénése du cortex surrénal des mam miféres. Comptes Rend. Ass. Anat., vol. 17, pp. 79-87.
1928 Les formations vesiculeuses dans les glandes endoerines.
Comptes Rend, Ass. Anat., vol. 23, pp, 69-75.
1933 Le eortex surrénal. Bull. Ass. Anat., No. 33.
Drerricu, A., AND H. Sirewunp 1926 Die Nebenniere und das chromaffine System. In: Handbuch der speziellen pathologischen Anatomie und Histologie, vol. 8, pp. 951-1089.
Dosneg, C., anp A. J. Davron 1941 Changes in the lipoid content of the adrenal gland of the rat under conditions of activity and rest. Anat. Rec., vol. 80, pp. 211-217.
Fixx, W. 1941 Histochemisehe Studien tiber Vitamin C und Plasmalogen an Ovarinm und Nebennierenrinde vom Meerschweinehen. Zeitschr. f. mikrosk.-anat. Forseh., vol. 50, pp. 558-589.
Gerscu, I., anp A, GroLLMAN 1939 The nature of the x-zone of the adrenal gland of the mouse, Anat. Ree., vol. 75, pp. 131-153.
GoormaGuticH, N. 1918 Contribution 4 1’étude du fonetionnement de la capsule surrénale humaine & 1’état normal et dans les états infeetieux en particulier dans les gaugrénes gazeuses. Arch. Méd. Exp., vol. 28, pp. 277-321. 414 PETER GRUENWALD
GooRMAGHTIGH, N. 1921 Organogénése de las capsule surrénale et du plexus eoeliaque. Arch. de Biol., vol. 31, pp. 83-172.
Groat, R. A. 1943 Adrenocortical-like tissue in the ovaries of the adrenaleetomized ground squirrel (Citellus tridecemlineatus). Endocrinology, vol. 32, pp. 488-492.
1944 Formation and growth of adrenocortical-like tissue in the
ovaries of the adrenalectomized ground squirrel, Anat, Ree., vol, 89, pp. 33-41.
GRUENWALD, P. 1942a Common traits in development and structure of the organs originating from the ecoelomie wall. J. Morphol., vol. 70, pp. 353-387.
GRUENWALD, P. 1942b The modes of origin of dvystopic tissues, with special reference to the problem of hypernephroma. J. Urol., vol. 48, pp. 244-251.
CGRUENWALD, P., AnD W. M. Konikov 1944 Cell replacement and its relation to the zona glomerulosa in the adrenal cortex of mammals. Anat. Ree., vol. 89, pp. 1-21.
Gutrman, P. H. 1930 Addison’s disease. Arch. Pathol., vol. 10, pp. 742-785, 895-935,
Hecker, L. 1942 Untersuchungen tiber das Vorkommen von Vitamin C in der Nebenniere des Mensehen. Zeitschr. f. mikrosk.-anat,. Forsch., vol. 52, pp. 393-417.
Herr, J. 1925 Ein Beitrag zur Histogenese der menschlichen Nebenniere. Zeitsehy, f. mikrosk.-anat. Forseh., vol. 3, pp. 179-282.
Hint, R. T. 1946 Life sustaining adrenal cortical function of the mouse ovary. Anat. Ree., vol. 94, p. 470.
Hoerr, N. L. 1931 The cells of the suprarenal cortex in the guinea pig. Their reaction to injury and their replacement. Am. J. Anat., vol. 48, pp. 139-197.
Keener, M. FP. L., anp BE. E, Hewer 1927 Observations on the development of the human suprarenal gland. J. Anat., vol. 61, pp. 802-324.
Konno, 8. 1925 Zur vergleichenden Histologic und Embryologie der Nebenniere der Sduger und des Menschen. Zeitsehr. f. Anat. u. Entwicklungsgesch., vol. 77, pp. 419-480.
Koimer, W. 1918 Zur vergleichenden Histologie, Zytologie und Entwicklungsgeschichte der Saugernebenniere. Arch, f. mikrosk. Anat., vol. 91, pp. 1-139.
Kovacs, W. 1928 Zur Nebennierenpathologie. Beitr. z, pathol. Anat., vol. 79, pp. 213-267.
Kroczex, H. 1941 Die Plasmalreaktion in der Nebennierenrinde des Meerschweinchens unter experimentellen Bedingungen. Zeitschr. f. mikrosk.anat, Forseh., vol. 50, pp, 511-521.
Lewis, R. W., ano A. M. PAPPENHIEIMER 1916 <A study of the involutional changes which oceur in the adrenal eortex during infancy. J. Med. Research, vol. 34, pp. 81-94.
Mutton, M. P. 1903 Divisions nucléaires et rédle germinatif de Ja couche glomérulaire des capsules surrénales du cohbave. Comptes Rend. Soe. Biol., vol, 55, pp. 592-595, DEVELOPMENT OF THE ADRENAL CORTEX 415
Potirzer, G. 1936 Ueber die Frithentwicklung der Nebennierenrinde beim Menschen. Zeitschr. f. Anat. u. Entwicklungsgesch., vol. 106, pp. 40-48.
Popper, H. 1941 Histologic distribution of vitamin A in human organs under normal and under pathologic conditions. Arch. Pathol., vol. 31, pp. 766— 802.
Ricu, A. R. 1944 <A peeuliar type of adrenal cortical damage associated with acute infections, and its possible relation to circulatory collapse. Bull. Johns Hopkins Hosp., vol. 74, pp. 1-15.
Rocorr, J. M. 1944 Experimental pathology and physiology of the adrenal cortex. Production of Addison’s disease in laboratory animals. Arch. Pathol., vol. 38, pp. 392-409.
Sarason, E, L. 1943 Morphologic changes in the rat’s adrenal cortex under various experimental conditions. Arch. Pathol., vol. 35, pp. 373-390.
———-—— 1943 Adrenal cortex in systemic disease. A morphologie study. Arch, Int, Med., vol. 71, pp. 702-712.
Srecksén, A. 1902 Befund von ‘‘Adenom’’ knétehen an Nebennieren und von accessorische Nebennieren bei Erwachsenen. Arb. a. d. Gebiete d. pathol. Anat. a. d. pathol.-anat. Inst. Tiibingen, vol. 3, pp. 253-260.
Tuomas, E. 1911 Ueber die Nebennicre des Kindes und ibre Veranderungen bei Infektionskrankheiten, Beitr. z. pathol. Anat., vol. 50, pp. 283-316.
Tonctti, E. 1941 Hormonal gesteuerte Transformationsfelder in der Nebennierenrinde? Zeitschr. f. mikrosk.-anat. Forsch., vol. 50, pp. 495-501.
---—-——— 19424 Zur Histophysiologie der Nebennierenrinde: Bau und Histochemie bei der Atrophie des Organs nach Hypophysektomie. Zeitschr. f. mikrosk.-anat. Forsch., vol. 51, pp. 346-392.
— — 1942b Die Umbauvorginge in den Transformationsfeldern der Nebennicrenrinde als Grundlage der Beurteilung der Nebennierenarbeit. Zeitsehr. f. mikrosk.-anat. Forseh., vol. 52, pp. 32-86.
Uorina, U. U. 1940 The early embryological development of the fetal and permanent adrenal cortex in man. Anat. Rec., vol. 76, pp. 183-203.
Wakrine, H. 1935 The development of the adrenal gland of the mouse. Quart. J, Microse. Sci., vol. 78, pp. 329-366.
WHITEHEAD, R. 1943 The growth of the adrenal cortex in the guinea pig. J. Pathol. and Bact., vol. 55, p. 392.
WILLIAMS, R. G. 1945 The characteristics and behavior of living cells in autogenous grafts of adrenal cortex in rabbits. Am. J. Anat., vol. 77, pp. 53-79.
Worton, R. M., anp R. L. ZwemeR 1943 A study of the cytogenesis of corticoadrenal cells in the cat. Anat. Ree., vol. 86, pp. 409-416.
Zwemer, R. L., R. M. Wotton anp M. G. Norkvs 1938 <A study of corticoadrenal cells, Anat. Ree., vol. 72, pp. 249-263.
PLATE i EXPLANATION OF FIGURES
1-3 Adrenals of human embryos of 67, 145 and 170mm crown-heel Jength, respectively, 1 and 2, Gomori’s silver impregnation; 3, hematoxylin-eosin. In contrast to the early stage (fig. 1), figure 2 shows a peripheral layer of the cortex in which individual cells are not surrounded by lattice fibers. Figure 3 shows numerous cavities and cells of small size in this peripheral layer.
4+ Adrenal of a 2-day-old female infant with a zona glomerulosa consisting of large units which appear on seetion as arches. Gomori’s silver impregnation,
5 Adrenal of a 15-day-old male infant. Gomori’s silver impregnation. The large units of the zona glomerulosa are being subdivided into narrow cords by new lattice fiber septa.
6-7 Adrenal of a 22-month-old infant. 6, Gomori’s silver impregnation; 7, azan stain. The zona glomerulosa contains very wide arches. Figure 7 shows the transition of small cells in the zona glomerulosa (right half of figure) into large cells (lett half).
8 Adrenal of a 6-month-cld female infant. Azan stain. The zona glomerulosa contains many cavities.
9 Adrenal of a boy of 1 vear and 7 months. Gomori’s silver impregnation. The zona glomerulosa consists of thick cell cords, and is almost completely separated from the deeper Javers bv lattice fibers.
DEVELOPMENT OF THE ADRENAL CORTEX PLATE 1
PLATE 2 EXPLANATION OF FIGURES
Accessory cortical nodules
10 Adrenal of a nodules. Azan stain.
11 Adrenal of a 24-month-old infant (same as in figs. 6 and 7) with a mass of accessory cortical tissue in the capsule, connecting at one point with the main organ. Van Gieson stain.
12 Accessory nodule in the adrenal capsule of a 4-month-old male infant. Hematoxyvlin-eosin stain, The capsule of the nodule contains cortical cell cords in tangential arrangement between its fiber layers (arroy
13. Nodule in the epididymis of a 2-month-old infant. Hemutoxylin-eosin stain. There are similar cell cords in the capsule as in figure 12. The right side of the figure shows the region ef the degenerating x-zone.
‘month-old male infant. showing numerous accessory
PLATE 3 EXPLANATION OF FIGURES
14 Adrenal of a boy of 2 years and 8 months. Gomori’s silver impregnation. Many glomerulosa cords have tapering distal ends which bend and approach the direction of the capsule.
15-16 Adrenal of a 6-year-old girl, 15, Van Gieson stain; 16, Gomori’s silver impregnation, higher magnification. There is a layer of thin cortical cell cords between the eapsule and large arches usually seen in the zona glomerulosa.
17 Adrenal of a 4-vear-old boy. Azan stain, The peculiar arrangement of a group of cell cords with the cortex is explained in the text.
18 Adrenal of a 59-year-old man. Gomori’s silver impregnation. For an explanation of the course of the cell cords in the cortex, see the text.